[R-sig-Geo] Dividing World Map

[Roger Bivand]

On Mon, 1 Oct 2012, Massimiliano Ruocco wrote:

> Thanks for the answer
> Yes, should be fine for me. I don't need polar areas. So shoud be fine to go 
> from -80 to 80 for lat.
> So an approximative idea to proceed could be the following:
>
> 1) scan all the UTM zone
> 2) divide the utm zone in quad of 10kmX10km
>
> Just a question...i am not sure i really understand the UTM coord. Is each 
> zone a 500kmX500km quad?

Maybe drop in on Jan Ketil Rød:

http://www.svt.ntnu.no/geo/jan.rod/

as you seem to need local help. He also wrote a good introductory book to 
these subjects:

Verktøy for å beskrive verden. Statistikk, kart og bilder.

if you read Norwegian.

Roger

>
> THanks again
>
>
>
> On 10/01/2012 10:05 AM, Barry Rowlingson wrote:
>> On Mon, Oct 1, 2012 at 8:09 AM, Massimiliano Ruocco <ruocco at idi.ntnu.no> 
>> wrote:
>>> Dear all,
>>> i would like to divide the world map in grid of equal dimension (example
>>> quad of 10kmx10km). Is it correct to use the UTM coord instead of
>>> lat/lon ones? Is there any drawback on doing that?
>>> 
>>> IS there any existing routine doing it?
>>   Since we figured out that the earth isn't flat we've known its
>> impossible. You can't divide a sphere into a grid like that - what are
>> you going to do at the poles?
>>
>>   You can project small parts to coordinates such as UTM and divide
>> into a regular square grid in those coordinates, but even those aren't
>> perfect, and get less perfect nearer the poles, at which point they
>> are completely wrong...
>> 
>> I think UTM zones typically have a max latitude they should be used
>> at, so as long as you don't care about polar bears and penguins you
>> should be okay...
>> 
>> Barry
>
>
>

-- 
Roger Bivand
Department of Economics, NHH Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no